Epilepsy patients who are candidates for surgical resection of a brain lesion must have some form of functional mapping to determine if the lesion can be removed without creating a functional deficit. The current procedures for mapping language in these patients are spatially limited (e.g. hemispheric mapping with Wada testing) or highly invasive (cortical stimulation using subdural electrodes). Functional MR imaging (fMRI) has the potential to replace, or significantly enhance, the current methods used in Neurosurgical planning. It is noninvasive and has been shown to be able to localize cortical activity. However, the standard echo planar imaging fMR1 methods, based on the BOLD activation response, suffer from low spatial resolution, and sensitivity to static Bo magnetic field inhomogeneities which lead to image distortions and low signal intensity. These problems severely limit the ability of fMRI to localize brain activity for Neurosurgical planning. This project has 3 aims. First, it is designed to improve fMRI methodology. Our earlier work has identified the need to develop single shot approaches to maximize statistical power in fMRI studies, and thus the methodology developments will focus on single shot approaches to image distortion and signal loss, including the investigation of asymmetric spin echo EPI and dynamic shimming. This innovative program will allow for highly robust functional localization required for surgical planning, but the methods to be developed may be applied to any fMRI study. Secondly, we will continue our work on validating the fMRI activation detected through comparison with Wada testing and cortical stimulation. A battery of fMRI language paradigms, parallel to out-of-magnet behavioral studies, have been designed to address the specific deficits obtained in temporal lobe epilepsy patients and these will be applied to both control subjects and patients with intractable epilepsy who are candidates for surgery. Patients will be imaged and tested behaviorally both pre and post- surgery. Through such studies the ability of fMRI to predict surgical outcome will be evaluated. Finally, we will examine the relationship between disease states (mesial temporal sclerosis, cortical malformations, and tumors) and language reorganization both acutely (pre- and post- surgery) and chronically by relating the language organization and performance prior to surgery, to the type, location, and age at onset of epilepsy. Such a study will provide evidence not only of language reorganization but also of the specific cortical regions that reorganize, and the impact this reorganization has on performance and surgical outcome.